The present invention relates to continuous availability between sites at unlimited distances, and more specifically, to the management and synchronization of all workloads in a multisite continuous workload availability computing environment.
The present invention relates to continuous availability between sites at unlimited distances, and more specifically, to maintaining a two-site configuration in a multisite continuous availability computing environment.
In the past, some computer availability and disaster recovery solutions were limited to a maximum distance between sites. Other past solutions required starting systems, applications, and supporting infrastructure on the backup site that could in some cases take several hours to restart. Some past solutions additionally required modifications to software applications, such as database servers, and hardware, such as routers and switches, in order to implement various disaster recovery and continuous availability functions, resulting in relatively high implementation cost. Some past solutions operated at a site level, rather than at a workload level.
These issues have been substantially addressed by continuous availability solutions between sites at unlimited distances. However, when a primary or secondary continuous availability computer processing site becomes unavailable, for example, due to an unplanned outage or during planned maintenance at one of the sites, the remaining site may run in a single-site configuration, without continuous workload availability, throughout the duration of the primary or secondary site unavailability, until such time as the site becomes available once again. This scenario may expose some workloads to a higher level of risk during a substantial period of time.
In addition, some solutions provide continuous workload availability only with regard to routable, online transaction processing (OLTP) applications and related data objects. Non-OLTP workloads, or “batch” workloads, may not be supported. In some cases, unsupported workloads may target the same data objects as supported workloads. However, a period of unavailability of a primary site may result in the loss or unavailability of data related to unsupported workloads, such as intermediate data files and job scheduler states. In addition, data related to unsupported workloads may be inconsistent with duplicate data related to supported workloads.